Secondary terms in the counting functions of quartic fields II
Arul Shankar, Jacob Tsimerman · 2025
Read Secondary terms in the counting functions of quartic fields II by Arul Shankar, Jacob Tsimerman (2025) in the Tower of Babel, a free online library of public-domain works preserved in clean, searchable text. No sign-up. The full text is served as a plain-text file at a predictable URL, where text is present: https://antarctic-labs.com/tower-of-babel/arxiv-2508-08527.txt.
Introduction
Arul Shankar and Jacob Tsimerman refine counts of quartic number fields in a 2025 maths paper. Quartic fields are number systems built from roots of degree-four equations, sorted by a size gauge called the discriminant. Precise counts with local rules imposed have long been a core aim.
The smoothed count now resolves into two main terms plus an error that shrinks by a power. The same holds for quartic rings, weighed by their cubic partners. A side result caps how wild the linked zeta functions can grow at a key point. The work is part two of a series on such counts.
Catalog record
| CREATOR | Arul Shankar, Jacob Tsimerman |
|---|---|
| YEAR | 2025 |
| COLLECTION | PAPERS |
| FORMAT | TXT |
| FILE SIZE | 121 KB |
| VERSION | 1.0 |
| LICENSE | http://creativecommons.org/licenses/by/4.0/ |
| RIGHTS | CC BY |
| SOURCE | arXiv |
Source: https://arxiv.org/abs/2508-08527
Browse the full library · About the Tower of Babel
Full text
The opening of the work, from the library's hosted text:
Secondary terms in the counting functions of quartic fields II Arul Shankar, Jacob Tsimerman Abstract We determine the smoothed counts of S 4 S_{4} -quartic fields with bounded discriminant, satisfying any finite specified set of local conditions, as the sum of two main terms with a power saving error term. We also prove an analogous result for quartic rings (weighted by the number of cubic resolvents), deducing as a consequence that the Shintani zeta functions associated to the prehomogeneous vector space ℂ 2 ⊗ Sym 2 ( ℂ 3 ) {\mathbb{C}}^{2}\otimes{\rm Sym}^{2}({\mathbb{C}}^{3}) have at most a simple pole at s = 5 / 6 s=5/6 . 1 Introduction In the first article [ 17 ] of this two-part series, we carried out the “smooth count” of quartic S_{4} -fields, in certain natural families, with two main terms of sizes X and X^{5/6} . Our main result there was restricted to what we called S_{4} -families , which are families of quartic fields, satisfying a finite set of local conditions, which ensures that every field in the family is a quartic S_{4} -field. The purpose of this article is to remove this restriction, and provide a smooth count for families of quartic fields satisfying any finite set of local conditions. 1 1 1 In our main result in this article, we evaluate smoothed counts of S_{4} -fields in all congruence families of quartic fields. Since asymptotics for non- S_{4} -quartic fields are already known (with an error term of o(X^{5/6}) ), our result can be used to deduce smoothed counts for the family of all quartic fields in any congruence family with two main terms. To state our main result, we need some notation. Let S be a finite set of places, and let \Sigma=(\Sigma_{v})_{v\in S} be a finite collection of local specifications for quartic fields , where for each v\in S , \Sigma_{v} is a finite set of étale quartic extensions of {\mathbb{Q}}_{v} . We let F(\Sigma) denote the family of S_{4} -quartic fields K such that K\otimes{\mathbb{Q}}_{v}\in\Sigma_{v} for every v\in S . Let \psi:{\mathbb{R}}_{\geq 0}\to{\mathbb{R}}_{\geq 0} be a smooth function with compact support, and define the “smooth count” of quartic fields in F(\Sigma) by N_{\Sigma}(\psi,X):=\sum_{K\in F(\Sigma)}\psi\Bigl(\frac{|\Delta(K)|}{X}\Bigr). (1) For a principal ideal domain R , Bhargava’s landmark work [ 3 ] gives a parametrization of the set of quartic rings Q over R along with a cubic resolvent ring C of Q . Given Q and C , there is a natural quadratic resolvent map r:Q/R\to C/R . In this paper we work with triples (Q,C,r) . Given such a triple over R , and given an element x\in(C/R)^{\vee} , we obtain (by composition with r ) a quadratic form Q/R\to R , which we denote by r_{x} . Specialize to the case R={\mathbb{Q}}_{p} for some p . For an étale quartic extension K of {\mathbb{Q}}_{p} , Bhargava proves that the ring of integers {\mathcal{O}}_{K} of K has a unique cubic resolvent C_{K} . We say that ({\mathcal{O}}_{K},C_{K},r_{K}) is the triple corresponding to K , we denote the quadratic form corresponding to an element x in (C_{K}/{\mathbb{Z}}_{p})^{\vee} by r_{K,x} , and the set of primitive elements in (C_{K}/Z_{p})^{\vee} by (C_{K}/{\mathbb{Z}}_{p})^{\vee}_{\rm prim} . Define the constants \mathcal{M}:=\frac{2^{5/3}\Gamma(1/6)\Gamma(1/2)}{\sqrt{3}\pi\Gamma(2/3)};\;\;\mathcal{M}_{i}:=\left\{\begin{array}[]{rcl}\mathcal{M}&{\rm if}&i\in\{0,2\};\\ \sqrt{3}\cdot\mathcal{M}&{\rm if}&i=1;\end{array}\right.\;\mathcal{M}_{i}^{\prime}:=\left\{\begin{array}[]{rcl}\mathcal{M}&{\rm if}&i=0;\\ \sqrt{3}\cdot\mathcal{M}&{\rm if}&i=1;\\ \displaystyle\frac{\mathcal{M}}{3}&{\rm if}&i=2.\end{array}\right. (2) Then we have the following result generalizing [ 17 , Theorem 2] . Theorem 1 . Let \Sigma be any finite collection of local specifications and let F(\Sigma) be the corresponding family of S_{4} -quartic fields. Let \psi:{\mathbb{R}}_{\geq 0}\to{\mathbb{R}}_{\geq 0} be a smooth function with compact support. Then N_{\Sigma}(\psi,X)=C_{1}(\Sigma)\widetilde{\psi}(1)\cdot X+C_{5/6}(\Sigma)\widetilde{\psi}(5/6)\cdot X^{5/6}+O(X^{13/16+o(1)}), where C_{1}(\Sigma):=\frac{1}{2}\Bigl(\sum_{K\in\Sigma_{\infty}}\frac{1}{\#{\rm Aut}(K)}\Bigr)\prod_{p}\Bigl(\sum_{K\in\Sigma_{p}}\frac{|\Delta(K)|_{p}}{\#{\rm Aut}(K)}\Bigr)\Bigl(1-\frac{1}{p}\Bigr), (3) and \displaystyle C_{5/6}(\Sigma) \displaystyle:=\frac{\pi}{8}\Bigl(\mathcal{M}_{\Sigma}\cdot\zeta(1/3)\prod_{p}\Bigl(1-\frac{1}{p^{1/3}}\Bigr)\sum_{K\in\Sigma_{p}}\frac{|\Delta(K)|_{p}}{\#{\rm Aut}(K)}\cdot\int_{x\in(C_{K}/{\mathbb{Z}}_{p})^{\vee}_{{\rm prim}}}|\det{r_{K,x}}|_{p}^{-2/3}dx \displaystyle\;\;\;+\;\mathcal{M}_{\Sigma}^{\prime}\cdot\zeta(2/3)\prod_{p}\Bigl(1-\frac{1}{p^{2/3}}\Bigr)\sum_{K\in\Sigma_{p}}\frac{|\Delta(K)|_{p}}{\#{\rm Aut}(K)}\cdot\int_{x\in(C_{K}/{\mathbb{Z}}_{p})^{\vee}_{{\rm prim}}}\epsilon_{p}(r_{x})|\det{r_{K,x}}|_{p}^{-2/3}dx\Bigr). Above, ({\mathcal{O}}_{K},C_{K},r_{K}) is the triple corresponding to K , \epsilon_{p} denotes the Hasse invariant, and \mathcal{M}_{\Sigma}=\sum_{K={\mathbb{R}}^{4-2i}\times{\mathbb{C}}^{i}\in\Sigma_{\infty}}\frac{\mathcal{M}_{i}}{|{\rm Aut}(K)|};\quad\quad\mathcal{M}_{\Sigma}^{\prime}=\sum_{K={\mathbb{R}}^{4-2i}\times{\mathbb{C}}^{i}\in\Sigma_{\infty}}\frac{\mathcal{M}_{i}^{\prime}}{|{\rm Aut}(K)|}. See [ 17 , §9.6] for the explicit values in the product for various \Sigma . Cohen–Diaz y Diaz–Olivier [ 9 , Corollary 6.1] prove that the number of D_{4} -quartic fields, with discriminant bounded by X , grows like a constant times X up to an error of O(X^{3/4+o(1)}) . Durlanik [ 11 ] proves versions of this result, where the count is smoothed, and the family of all D_{4} -quartic fields is replaced by a family satisfying finitely many local conditions. Meanwhile, the number of V_{4} -, C_{4} -, and C_{2}\times C_{2} -fields with discriminant less than X is bounded by O(X^{1/2+o(1)}) by work of Baily [ 2 ] , and the number of A_{4} -quartic fields with discriminant less than X is bounded by O(X^{.778...}) in [ 5 , Theorem 1.4] . Therefore, Theorem 1 can be used to obtain the smooth count of all quartic fields (or indeed, quartic fields satisfying any finite set of local conditions) with bounded discriminant. In [ 17 , Theorem 3] , we prove a smoothed counting results for S_{4} -families of quartic rings (along with their resolvents). In this article, we generalize that result to any congruence family. We now precisely define the set of families that we can handle: \Lambda is said to be a finite collection of local specifications for quartic rings if \Lambda=(\Lambda_{v})_{v\in S} , where for all v in a finite set of places S , the set \Lambda_{v} consists of a finite set of quartic triples over {\mathbb{Z}}_{v} with non-zero discriminant. We define R(\Lambda) to be the set of quartic triples (Q,C,r) over {\mathbb{Z}} , whose base change to {\mathbb{Z}}_{v} lies in \Lambda_{v} for every v\in S , and such that Q is an order in an S_{4} -quartic field. Let \psi:{\mathbb{R}}_{\geq 0}\to{\mathbb{R}} be a smooth function with compact support. We define the smoothed count of quartic triples in R(\Lambda) analogously to ( 36 ): N_{\Lambda}(\psi,X):=\sum_{(Q,C,r)\in R(\Lambda)}\psi\Bigl(\frac{|\Delta(Q)|}{X}\Bigr). (4) Then we have the following result: Theorem 2 . Let \Lambda be any finite collection of local specifications for quartic rings and let R(\Lambda) be the corresponding family of quartic triples. Let \psi:{\mathbb{R}}_{\geq 0}\to{\mathbb{R}}_{\geq 0} be a smooth function with compact support. Then N_{\Lambda}(\psi,X)=C_{1}(\Lambda)\widetilde{\psi}(1)\cdot X+C_{5/6}(\Lambda)\widetilde{\psi}(5/6)\cdot X^{5/6}+O(X^{3/4}\log X), where C_{1}(\Lambda)=\frac{1}{2}\Bigl(\sum_{K\in\Lambda_{\infty}}\frac{1}{\#{\rm Aut}(K)}\Bigr)\prod_{p}\Bigl(\sum_{(Q,C,r)\in\Lambda_{p}}\frac{|\Delta(Q)|_{p}}{\#{\rm Aut}((Q,C,r)}\Bigr)\Bigl(1-\frac{1}{p}\Bigr), and \displaystyle C_{5/6}(\Lambda) \displaystyle=\frac{\pi}{8}\Bigl(\mathcal{M}_{\Lambda}\zeta(1/3)\prod_{p}\Bigl(1-\frac{1}{p^{1/3}}\Bigr)\cdot\sum_{(Q,C,r)\in\Lambda_{p}}\frac{|\Delta(Q)|_{p}}{\#{\rm Aut}(Q,C,r)}\cdot\int_{x\in(C/{\mathbb{Z}}_{p})^{\vee}_{{\rm prim}}}|\det{r_{x}}|_{p}^{-2/3}dx \displaystyle\;\;\;+\;\mathcal{M}_{\Lambda}^{\prime}\zeta(2/3)\prod_{p}\Bigl(1-\frac{1}{p^{2/3}}\Bigr)\cdot\sum_{(Q,C,r)\in\Lambda_{p}}\frac{|\Delta(Q)|_{p}}{\#{\rm Aut}(Q,C,r)}\cdot\int_{x\in(C/{\mathbb{Z}}_{p})^{\vee}_{{\rm prim}}}\epsilon_{p}(r_{x})|\det{r_{x}}|_{p}^{-2/3}dx\Bigr). Above, notation is as in Theorem 1 . For a ring R we define V(R)=R^{2}\otimes{\rm Sym}^{2}(R^{3}) . The group {\rm GL}_{2}\times{\rm GL}_{3} acts naturally on V , and this representation is prehomogeneous (the action over {\mathbb{C}} has a unique open orbit). Foundational work of Sato–Shintani [ 15 ] associates Shintani zeta functions (defined in §2.3) to prehomogeneous representations, and they develop general theory proving that these representations have meromorphic continuation to the complex plane with constrained orders and locations of the poles. For the specific case of (V,G) , it follows that the corresponding Shintani zeta functions have possible poles of order at most four at s=1 , and of order at most two at s=5/6 and s=3/4 . From Bhargava’s landmark counting results [ 4 ] , it follows that the possible pole at s=1 has order at most 2 . Using our counting results, we prove an analogous result about the possible pole at s=5/6 . Theorem 3 . Let \phi:V({\mathbb{Z}})\to{\mathbb{R}} be a G({\mathbb{Z}}) -invariant function, defined by congruence conditions modulo a positive integer. Then for i\in\{0,1,2\} , the Shintani zeta functions \xi_{i}(\phi;s) ( defined in